Da-Chief
07-15-2008, 11:30
http://i.dslr.net/urls/41/3341.gif (http://www.dslreports.com/shownews/Dont-Fear-The-Bandwidth-Apocalypse-96115)
A good rule of thumb: when someone claims the Internet is facing bandwidth armageddon, it's usually because they're in the business of designing and selling traffic shaping hardware (http://www.sandvine.com/), trying to justify new and frequently unjustifiable pricing models (http://www.corpsman.com/shownews/95759), or trying to sell some idea to federal regulators (usually less regulation and legal oversight (http://www.forbes.com/2007/01/30/info-traffic-jams-oped-cx_pk_0131network.html?partner=yahootix)). The guys actually working in the network operation centers will generally tell you that congestion can almost always be handled with smart design and capacity.
Last week the National Cable and Telecom Association (NCTA) was busy trying to lobby the FCC, which has been investigating exactly what sort of network management should be allowed, and how it should be disclosed to consumers. The NCTA argued that the use of deep packet inspection hardware was absolutely necessary on cable networks. Without such technology (the likes of which is being used to throttle Comcast P2P users), the NCTA claims that the Internet would all but collapse. From a series (http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi?native_or_pdf=pdf&id_document=6520033981) of (http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi?native_or_pdf=pdf&id_document=6520033982) letters (http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi?native_or_pdf=pdf&id_document=6520033983) sent to the FCC last Friday:In particular, (Insight CEO Mike) Willner described how, in the absence of network management, the usage of P2P services by a very small number of a cable system's high-speed Internet customers can cause substantial (and sometimes complete) congestion of the system's upload capacity. As a result, service for the system's high-speed Internet customers using the Internet for other purposes (such as e-mail, web browsing, e-shopping, streaming music and video, etc.) would be degraded.
http://www.corpsman.com/quote_left_white.gifnetwork upgrades that are intended to enhance the speed and quality of Internet access would, in the absence of network management, only exacerbate this problem.http://www.corpsman.com/quote_right_white.gif
-NCTA
The problem is, any claim of "complete congestion" is lobbyist hyperbole, again highlighting the chasm between lobbyists and real technicians. Networking and protocol specialist Robb Topolski should know -- he first discovered Comcast's use of Sandvine to throttle upstream capacity in May of 2007 (http://www.corpsman.com/forum/r18323368-Comcast-is-using-Sandvine-to-manage-P2P-Connections). It was his findings in our forums that led to the FCC's investigation of the cable company.
"Complete congestion is a technical fantasy which only exists in the minds of people who do not understand TCP congestion control and how Additive Increase/Multiplicative Decrease (AIMD) works in TCP Congestion avoidance works, he says. "AIMD allows a linear growth of bandwidth utilization until loss occurs, at which time an exponential reduction takes place. This slow-start, fast-fallback ensures congestion cannot cause gridlock."
In other words, total gridlock does not happen because it cannot happen, yet there's no shortage of people suggesting it's inevitable unless party X (ISPs, lobbyists, hardware salesmen) get what they want (less regulation, per-byte billing, a new Audi). In the letter to the FCC, the NCTA goes on to suggest that capacity upgrades wouldn't help -- and in fact would hurt:As Mr. Willner pointed out, network upgrades that are intended to enhance the speed and quality of Internet access would, in the absence of network management, only exacerbate this problem because P2P users around the world seek to retrieve files from computers on systems with the fastest upload speeds.
Topolski says the NCTA lobbyists are intentionally confusing providing network upgrades with providing faster upload speeds. "In an ISP, such as Insight's network, a network upgrade ought to be performed if the network is routinely experiencing congestion," he notes. "This does not mean that individual cable modems ought to be provisioned with higher speeds, but that the shared pool of uplink bandwidth that they share ought to be increased to reduce the occurrence of congestion."
So the NCTA is both falsely inflating the threat of "complete" congestion, while lying about whether increasing capacity would actually help. Why? The group is protecting the cable industry's right to provide less product for more money using dubious practices (Comcast and Cox's throttling of upstream bandwidth using forged TCP packets), while protecting possible DPI revenue streams like behavioral advertising (http://www.corpsman.com/shownews/96006).
Topolski tells me it's also important to understand that the NCTA is intentionally conflating network management with deep packet inspection. The two are not synonymous. Topolski recently authored an interesting report (http://www.corpsman.com/shownews/95385) on how behavioral advertising and deep packet inspection technology from NebuAD has a number of nasty habits.
"Deep Packet Inspection devices capable of detecting what applications end points are using are both new and intrusive. They do not perform a task of Network management, they are performing tasks of Session and Application management. These are both new and inappropriate roles for Internet Service Providers."
Like most techs, Topolski doesn't oppose reasonable network management -- but so far ISPs have shown their use of DPI gear to be anything but reasonable, much like their defense of packet forgery and behavioral advertising. "The Internet did not grow to become a raging success without management, and it borders on ridiculous for the industry to claim that the invention of full-scale wire-speed DPI is the Saviour of the Internet," says Topolski.
More...
A good rule of thumb: when someone claims the Internet is facing bandwidth armageddon, it's usually because they're in the business of designing and selling traffic shaping hardware (http://www.sandvine.com/), trying to justify new and frequently unjustifiable pricing models (http://www.corpsman.com/shownews/95759), or trying to sell some idea to federal regulators (usually less regulation and legal oversight (http://www.forbes.com/2007/01/30/info-traffic-jams-oped-cx_pk_0131network.html?partner=yahootix)). The guys actually working in the network operation centers will generally tell you that congestion can almost always be handled with smart design and capacity.
Last week the National Cable and Telecom Association (NCTA) was busy trying to lobby the FCC, which has been investigating exactly what sort of network management should be allowed, and how it should be disclosed to consumers. The NCTA argued that the use of deep packet inspection hardware was absolutely necessary on cable networks. Without such technology (the likes of which is being used to throttle Comcast P2P users), the NCTA claims that the Internet would all but collapse. From a series (http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi?native_or_pdf=pdf&id_document=6520033981) of (http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi?native_or_pdf=pdf&id_document=6520033982) letters (http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi?native_or_pdf=pdf&id_document=6520033983) sent to the FCC last Friday:In particular, (Insight CEO Mike) Willner described how, in the absence of network management, the usage of P2P services by a very small number of a cable system's high-speed Internet customers can cause substantial (and sometimes complete) congestion of the system's upload capacity. As a result, service for the system's high-speed Internet customers using the Internet for other purposes (such as e-mail, web browsing, e-shopping, streaming music and video, etc.) would be degraded.
http://www.corpsman.com/quote_left_white.gifnetwork upgrades that are intended to enhance the speed and quality of Internet access would, in the absence of network management, only exacerbate this problem.http://www.corpsman.com/quote_right_white.gif
-NCTA
The problem is, any claim of "complete congestion" is lobbyist hyperbole, again highlighting the chasm between lobbyists and real technicians. Networking and protocol specialist Robb Topolski should know -- he first discovered Comcast's use of Sandvine to throttle upstream capacity in May of 2007 (http://www.corpsman.com/forum/r18323368-Comcast-is-using-Sandvine-to-manage-P2P-Connections). It was his findings in our forums that led to the FCC's investigation of the cable company.
"Complete congestion is a technical fantasy which only exists in the minds of people who do not understand TCP congestion control and how Additive Increase/Multiplicative Decrease (AIMD) works in TCP Congestion avoidance works, he says. "AIMD allows a linear growth of bandwidth utilization until loss occurs, at which time an exponential reduction takes place. This slow-start, fast-fallback ensures congestion cannot cause gridlock."
In other words, total gridlock does not happen because it cannot happen, yet there's no shortage of people suggesting it's inevitable unless party X (ISPs, lobbyists, hardware salesmen) get what they want (less regulation, per-byte billing, a new Audi). In the letter to the FCC, the NCTA goes on to suggest that capacity upgrades wouldn't help -- and in fact would hurt:As Mr. Willner pointed out, network upgrades that are intended to enhance the speed and quality of Internet access would, in the absence of network management, only exacerbate this problem because P2P users around the world seek to retrieve files from computers on systems with the fastest upload speeds.
Topolski says the NCTA lobbyists are intentionally confusing providing network upgrades with providing faster upload speeds. "In an ISP, such as Insight's network, a network upgrade ought to be performed if the network is routinely experiencing congestion," he notes. "This does not mean that individual cable modems ought to be provisioned with higher speeds, but that the shared pool of uplink bandwidth that they share ought to be increased to reduce the occurrence of congestion."
So the NCTA is both falsely inflating the threat of "complete" congestion, while lying about whether increasing capacity would actually help. Why? The group is protecting the cable industry's right to provide less product for more money using dubious practices (Comcast and Cox's throttling of upstream bandwidth using forged TCP packets), while protecting possible DPI revenue streams like behavioral advertising (http://www.corpsman.com/shownews/96006).
Topolski tells me it's also important to understand that the NCTA is intentionally conflating network management with deep packet inspection. The two are not synonymous. Topolski recently authored an interesting report (http://www.corpsman.com/shownews/95385) on how behavioral advertising and deep packet inspection technology from NebuAD has a number of nasty habits.
"Deep Packet Inspection devices capable of detecting what applications end points are using are both new and intrusive. They do not perform a task of Network management, they are performing tasks of Session and Application management. These are both new and inappropriate roles for Internet Service Providers."
Like most techs, Topolski doesn't oppose reasonable network management -- but so far ISPs have shown their use of DPI gear to be anything but reasonable, much like their defense of packet forgery and behavioral advertising. "The Internet did not grow to become a raging success without management, and it borders on ridiculous for the industry to claim that the invention of full-scale wire-speed DPI is the Saviour of the Internet," says Topolski.
More...